PhD Oral Exam - Mohammadreza Amini, Computer Science

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Modeling human motor behavior is central to Human-Computer Interaction (HCI), providing foundations for designing, optimizing, and evaluating interactive systems. In HCI, steering is among fundamental interactions, and refers to continuous movement along constrained paths, widely studied using Steering law. Although extensively adopted in 2D interfaces, modeling steering remains challenging in 3D Virtual Environments (VEs) due to the complex characteristics of 3D interactions, e.g., lack of physical support, added degree of freedom, and head-mounted displays' perceptual limitations, which is particularly important given the rapidly emerging applications of immersive environments.

This thesis empirically extends Steering law to 3D VEs through a layered approach. It first establishes a methodological foundation for modeling motor performance, considering factors like task design, model formulation, and evaluation metric. It then shows the effect of the steering experimental task, proposes a novel trajectory-aware effective throughput calculation, and examines how speed-accuracy strategies and collider-bounded trajectories shape steering performance and evaluation in 3D. Building on these, Steering law extensions are developed for distal steering beyond arm’s reach and steering through 3D tunnels with linearly changing widths, to better capture steering performance considering domain-specific 3D interaction constraints. Lastly, an application-oriented task demonstrates the robustness of the changing-width model in a practical medical scenario.

Together, these studies show that Steering law remains a valuable predictive model for 3D VEs when it is empirically modeled and validated within the targeted scope, deepening our understanding of 3D steering behavior while providing evidence-based implications for effectively designing and evaluating steering interactions in immersive environments.